ratus for crushing or concentration is usually connected with chlorination works. In this account it will be assumed that the raw ore has in all cases been finely pulverized.

Ores containing sulphurets or arseniurets are prepared by roasting. That this class must comprise most of the ores treated by chlorination is evident when it is considered that the presence of these compounds is the chief obstacle to successful amalgamation, and that the concentration of the tailings from amalgamating mills is practically a collection of the sulphurets and arseniurets which they contain.

The object of roasting is to convert the base metals into oxides that will not unite with the chlorine, and to leave the gold in a condition suitable for its chlorination. The latter object may be defeated by one of three causes. The gold in the ore may be in coarse particles, which chloridize too slowly for practical economy. For this reason, ore containing coarse gold is not treated by the chlorination process. Again, the gold may be alloyed with silver, the chloride of which is insoluble, and may form a coating upon the particles, preventing the complete chlorination of the gold. Küstel thinks that gold of very low fineness, containing from 40 to 50 per cent. of silver, will probably resist the chlorination, unless it is in the finest state of pulverization. In such a case, as indeed wherever there is silver to be extracted, some modification of the Plattner process, such as the lixiviation with chlorinated brine, must be employed. This will be alluded to hereafter. Finally, it is possible, though perhaps not demonstrated as yet, that other substances, such as oxide of iron, for example, may coat the gold and hinder the action of the chlorine. This evil (if it exists) is remedied by an addition of salt in roasting. I speak with some doubt on this point, though the efficacy of the use of salt in roasting some ores for chlorination is undoubted. But the nature of this benefit may be otherwise explained than by supposing it to consist in the removal of a coating from the gold. If the latter were the case, then, it seems to me, salt would always be necessary in the preliminary roasting; but this is not the case. Experiments in Colorado (see page 346 of my last report) have indicated that a coating is left upon gold in the roasting of auriferous sulphurets, that it is probably oxide of iron, and that it may be removed by the addition of salt toward the end of the roasting;* but while this coating may be sufficient to prevent close contact of gold and quicksilver, and so hinder amalgamation, it does not necessarily follow that it will prevent the action of chlorine gas.

Turning to consider the first object of roasting, the oxidation of the base metals, we find that it must be conducted with great skill and care to insure the success of the subsequent chlorination. The following

Of the beneficial effect of salt in the roasting of auriferous ores, preliminary to amalgamation, I entertain little doubt. The experiments in Colorado, here referred to, were those of Mr. Brückner, with his roasting-cylinders. They are, perhaps, corroborated by the unexpected results of the first Stetefeldt furnace, erected at the Twin River Mill, Nye County, Nevada, and tested upon silver ores. It was found that the pan amalgamation of the roasted ore yielded a small percentage of gold in the silver bullion, which had not been the case when ordinary reverberatories were used. In this case, however, salt was employed in the reverberatories as well as in the Stetefeldt furnace; and the only explanation I can at present suggest for the difference in results is, that the finely pulverized condition of the ore in the latter, and the complete access afforded the chloridization agents to each particle, effect a complete chloridizing, and allow an action upon fine particles of gold not secured in the ruder reverberatory process. The reverberatories used in chlorination works, however, differ from those in ordinary silver-mills, and the roasting is more careful and thorough. In this case, salt added in the late stages of the process may be more likely to act upon the supposed coating of the gold. But its true function under such circumstances is, I think, in its action upon magnesia, lead, and perhaps lime.

conditions are involved, together with others, less important, or less peculiarly characteristic of this process:

1. In all roasting operations a high initial temperature is likely to cause a sintering of the sulphurets, and a formation of matte, which cannot be oxidized without a new pulverization. This evil is particularly to be dreaded when galena is present. A low heat and diligent stirring are therefore required at the beginning of the process.

2. Under these conditions, the sulphur of the sulphurets is set free, combining with the oxygen of the air to sulphurous acid, which escapes in gaseous form, and to sulphuric acid, which unites with the oxidized metals (especially iron) to form sulphates.

3. The sulphate of iron (protoxide) is, as has been already remarked, a precipitant of gold from its chloride solution. Hence its presence in the roasted ore will defeat the object of chlorination and lixiviation; and it is therefore necessary to destroy the sulphates. This is accomplished by gradually increasing the temperature until a point is reached at which these combinations are dissociated. The formation and subsequent decomposition in the charge of arseniates is governed by similar laws.

4. Lime and magnesia, as well as lead, exert an unfavorable influence on the chlorination, after roasting. For some time after the process had been successfully employed in Grass Valley, the concentrated sulphurets from the Eureka mine, in that district, presented a mysterious obstacle to its application. Chlorine was absorbed by them after roasting; but it seemed to be wasted upon some other substance than the gold. Yet the oxidation in the furnaces appeared to be reasonably complete. Mr. Deetken, who was called into consultation, succeeded in overcoming the difficulty, and became the manager of the chlorination works of the company. According to his experiments, it seems probable that lime, magnesia, (and lead oxide ?) are attacked by the chlorine, forming hypochlorites, or oxychlorides and chlorides, after the manner of alkalies. Magnesia, at least, undoubtedly shares with the alkaline bases this property of combining with chlorine. The indifference of the ordinary metallic oxides may, perhaps, be less complete than has been usually supposed. In the case of the Eureka sulphurets the troublesome constituent appears to have been magnesia from the gangue or country rock. The cure was the addition of salt toward the close of the roasting, and at high temperature, by which means the magnesia (sulphate ?) appears to have been chloridized.

5. The presence of any soluble metallic salts is injurious, since they at once react upon the oxide of iron, and the latter precipitates the gold from its solution before lixiviation.

6. Chemists understand that metallic oxides, which do not readily, or do not at all, react with chlorine to form chlorides, may decompose readily with hydrochloric acid, since the latter contains hydrogen, which satisfies the oxygen of the base. To explain the matter rudely, (and according to the old-school formulas which are still current among metallurgists,) the equation R O+CI=R CI+O represents a reaction which practically does not take place, the affinity of the metal for chlorine not being sufficient to expel free oxygen. But in the reaction expressed by RO+HCI=RC1+HO, the affinity of the metal for chlorine is reënforced by that of hydrogen for oxygen, and the formation of a metallic chloride and of water simultaneously occurs. But this formation of any other soluble chloride than that of gold is to be avoided, because it leads, as I have indicated, to a premature precipitation of gold. Hence,

the chlorine gas employed in this process must be carefully freed from muriatic acid.

7. Muriatic or hydrochloric acid acts injuriously in another way, namely, when by reason of incomplete roasting the charge contains metallic sulphurets. Decomposing these, the acid produces chlorides, setting free sulphureted hydrogen gas, which is a precipitant of gold from its chloride solution.

8. The free chlorine is both annoying and destructive of health. Care should therefore be taken to ventilate the works thoroughly and to protect the workmen as much as possible against the inhalation of the gas.

The roasting is performed, as I have said, in a reverberatory furnace. This kind of furnace derives its name from the fact that the ore treated in it lies upon a hearth, over which is a low arched roof; and the flame from the fuel in the fire-place at one end, passing over a dividing wall called the fire-bridge, "reverberates" along the roof, and is reflected upon the charge. In a muffle-furnace the flame is not allowed to come in contact with the ore, but surrounds and heats the muffle or small oven containing the latter, while the actual oxidation is effected by fresh air introduced from outside. Naturally there is a loss of heat in this form, and the reverberatory, which gives in the case now under discussion equally satisfactory results, is preferred on account of the saving of fuel. The Gerstenhöfer or the Stetefeldt furnace may perhaps hereafter be applied to this use with still greater economy.

Küstel gives several examples, with descriptions and diagrams, illustrating different forms of reverberatories employed. They may be classified as single and double furnaces. The latter has two hearths, one over the other; and the roasting is begun upon the upper and finished upon the lower hearth. The furnace erected by Mr. Deetken at the Eureka works, which appears to be one of the best forms, has the lower hearth placed, not immediately under the upper, but in continuation of it, on a level 7 feet 10 inches lower. The two are connected by a step-flue. The upper hearth is 6 feet wide by 39 long, and has six working-doors on each side. Through these the charge is stirred and pushed along, as desulphurization advances. The lower hearth, immediately next the fire-place, is much shorter. Here the final roasting takes place, with the addition of salt. A draught through the whole is maintained by means of a chimney 25 feet high and 28 inches square in the clear. Mr. Küstel says of this furnace that it requires more room than the ordinary double furnace, but the work of stirring is less tiresome, since the roaster is not obliged to step constantly up and down. Another advantage is the extent of the upper hearth, which receives nine tons of ore without difficulty, whereas the charging of a furnace two (or even three or four) stories high is troublesome if not favored by sloping ground. It takes about twenty hours to finish the roasting of a charge of 2,000 pounds of sulphurets; but by employing a large and long furnace, such as is here described, over ten tons can be treated continuously, the latest charge receiving its preliminary while the earliest receives its final roasting. The capacity of such a furnace appears from the following brief description of the process, nearly as given by Mr. Küstel.

The heat in the lower hearth is always kept bright. One ton is roasted below, and about nine tons are spread on the upper long hearth. Two roasters are constantly at work, mainly at the separate hearths, but together, when required, at either. The ore on the upper hearth is pushed along as the process proceeds, until it arrives at the flue leading

down to the lower hearth.* At this point it contains oxides and sulphates, with a small portion of undecomposed sulphurets. The previous charge on the lower hearth having been withdrawn, the charge nearest the flue (one ton) is 'pushed down and spread upon the lower hearth. Here at a lively heat, and with active stirring at intervals, the base metals are converted into oxides in about eight hours, and the charge is finally withdrawn into an iron car. As soon as a charge is drawn into the lower hearth, the following charges are moved forward in succession, and space is thus left nearest the chimney, at the remote end of the upper hearth for a new charge of one ton of raw sulphurets. The capacity of the furnace, with two roasters constantly employed, (or four in twenty-four hours, is therefore three tons, while that of an ordinary single reverberatory, employing one man, (or two in twenty-four hours,) is but little more than one ton.

The roasted ore is removed in the iron car to a cooling floor. In Mr. Deetken's Eureka Works this floor is in front of the furnace, and very near, so that the removal can be effected directly.


The chlorine gas is prepared from peroxide of manganese, salt, and sulphuric acid, in a leaden vessel. The proportions for a charge of three tons of roasted sulphurets are given as follows:

Peroxide of manganese, (pulverized)
Common salt, according to quality..
Sulphuric acid, 66° B..........



30 to 40



The water, salt, and manganese are introduced first into the generator, which is covered with a curved lid, fitting in an annular water-joint. Through this cover two lead pipes communicate with the interior, that is, with the open space above the mixture. One is a safety-tube or funnel-tube, bent twice upon itself and terminating above in a funnel, through which the sulphuric acid is introduced. The other is the pipe conveying the chlorine to the vat. A vertical shaft or rod through the center of the cover carries a pair of arms, with teeth, used by revolution as a stirrer of the charge.

The sulphuric acid is introduced in successive small quantities, as needed to maintain a lively generation of gas. The formula of the reaction is, Mn O2+Na Cl+2SO3 =Mn OS O3+Na OSO3+Cl, [or, according to modern chemistry, (NaCl)2+(H2(SO4))2+MnO2=Mn (SO4)+Na2 (SO4)+ (H2O)2+Cl2. See Barker's Chemistry, par. 140.] After all the acid has been added, when the action flags, it may be stimulated for a while by a moderate fire under the generator. Instead of the ingredients above named, the chlorine may be produced from one part peroxide of manganese, two parts muriatic acid, and one part sulphuric acid, diluted with one part water.

The gas escaping from the generator is purified of muriatic acid by

*Mr. Küstel says the charge is exposed to the preparatory roasting for about twenty-four hours on the upper hearth; but I cannot understand how this can be the case, since the rate of its advance depends entirely upon the removal of the earlier charges, and this is regulated by the final roasting on the lower hearth. With regard to the latter, he says (section 59, p. 250) a ton can be drawn out every eight hours, that is, three tons in twenty-four hours. As the charges are introduced at the same rate on the upper hearth, it follows that if the upper hearth carries, as he says, nine tons, each of them must remain upon it 9 × 8=72 hours. It is my impression that the quantity of ore on the upper hearth is not so large as this.

"washing" it through an arrangement almost exactly like a pneumatic trough, by which it passes through a stratum of half or three-quarters of an inch of water. This absorbs the muriatic acid, and a proportion of the chlorine, (about 23 volumes.) Warm water takes up less chlorine, and even a saturated solution of chlorine will still absorb muriatic acid. From the purifier the gas is conducted to the vats or tanks.


After the roasted ore has become sufficiently cool it is dampened with 4 or 5 per cent. of water, which, it is claimed, facilitates the mechanical passage and the chemical activity of the chlorine. It is then sifted into the chlorination vat. The sieve used for this purpose need have no more than 7 to 8 meshes per linear inch.

The European authorities say that the vessels used for chlorination must not be of wood or metal, and recommend earthen pots or bottles.* This is troublesome and expensive, compared with the simple treatment in vats employed in this country. The reason for the usual prohibition of wood has been, I presume, the amount of gold solution which it would absorb; but Mr. Deetken has completely overcome this evil by the simple expedient of coating the inside with equal parts of pitch and tar. Thus, instead of small earthen pots, he is enabled to use large tanks, holding several tons of roasted ore. These are circular in form, and possess a false bottom, about 1 inch from the real one. Upon the false bottom, which consists of boards placed about one-eighth of an inch apart, and pierced with half-inch holes from 10 to 12 inches apart, is spread a layer of clean quartz, 1 to 2 inches thick. Any other indifferent rock will do, but not a rock containing magnesia or lime. This first layer of quartz is coarse; over it smaller pieces are laid, and so on, decreasing in size till a layer of sand covers the whole, forming thus a filter from 4 to 5 inches thick. This filter remains in the vat. Upon it the ore is sifted, when duly prepared for chlorination, and the cover is put on. This is nearly flat, and of wood. It is suspended by a chain attached to its center from above, and can thus be swung to its place promptly. The edge all around is luted with wheaten dough. A small hole in the cover is left open, to allow the escape of air and to serve as a means of observing the moment when the chlorine appears on the top of the ore. When this moment arrives the whole is closed and plugged

with dough.

Chlorine is now conducted into the ore and permitted to operate from twelve to eighteen hours. Leakages of gas from the apparatus may be detected by the odor, and by the formation of white fumes when ap proached with a glass rod previously dipped in ammonia.

The coarser the gold the longer the chlorination. After, as a maximum, eighteen hours, the cover is taken off and water is introduced. Usually, if the process has been effective, free chlorine has passed through the body of the ore, and makes its appearance as a green gas on the surface. When the gold is fine this may take place after twelve hours.


When the gas appearing on the surface indicates that the whole mass is permeated with chlorine, the cover is removed, and water is introduced until the surface of the charge is covered. Then a cock at the

* See Crookes and Röhrig's Kerl's Metallurgy, vol. i, p. 637.

« ForrigeFortsett »